Fuzzy-Genetic Classifier algorithm for bank's customers
Modern finical banks are running in complex and dynamic environment which may bring high uncertainty and risk to them. So the ability to intelligently collect, mange, and analyze information about customers is a key source of competitive advantage for an E-business. But the data base for any bank is too large, complex and incomprehensible to determine if the customer risk or default. This paper presents a new algorithm for extracting accurate and comprehensible rules from database via fuzzy genetic classifier by two methodologies fuzzy system and genetic algorithms in one algorithm. Proposed evolved system exhibits two important characteristics; first, each rule is obtained through an efficient genetic rule extraction method which adapts the parameters of the fuzzy sets in the premise space and determines the required features of the rule, further improve the interpretability of the obtained model. Second, evolve the obtained rule base through genetic algorithm. The cooperation system increases the classification performance and reach to max classification ratio in the earlier generations.
Keywords: fuzzy system, genetic algorithm, rule extraction, E-business
Download Full-Text
ABOUT THE AUTHORS
Rashed Mokhtar Elawady
Aziza Asim
Sara Mahmoud Sweidan
Rashed Mokhtar Elawady
Aziza Asim
Sara Mahmoud Sweidan